1
264
T H E JOURNAL OF INDUSTRIAL A N D ENGIlVEERING C H E M I S T R Y .
June, 19x0
tion or from distillation with steam, in such amounts the 1st molasses. I n the Ist,sugar it amounts to that the tests, quantitatively applied, can serve easily 12.8 parts CH,O per million as against 9 . 5 parts per to distinguish between products to which it has, million in 1st sugar from juice without added formand those to which it has not been added (Nos. 2 , I O aldehyde, or an excess of 3 . 3 parts per million. I n and 13, and Nos. I , 4, 7 and 12). From these acidi- the first molasses it amounts to 31.2 parts per milfied formaldehyde solutions, the formaldehyde comes lion as against 6 . I parts per million in 1st molasses off most copiously in the first fractions. When not from juice without added formaldehyde, or an excess acidified this is less marked and the total recovery of 2 5 . I parts per million. The molasses is thus found is less in the same total volume of distillate (cf. Nos. to be as strongly charged with this excess of formalde3 and 9). When made alkaline with NaOH, as we hyde as was the juice from which it was made. Asdid in some preliminary trials, the tendency to hold suming that the yields of 1st sugar and 1st molasses back formaldehyde in the sugar solution was many ,are 9 . 5 per cent. and 6 per cent., respectively, of the fold greater. juice, then the amounts of formaldehyde retained b y Coming now to the main question which concerned them are about 1.2 per cent. and 6 per cent., respecus, viz., Does formaldehyde added to juice in the tively, of that added to the juice, or a total of 7 . 2 sugar-house all boil away or does it persist in any per cent., which, as far as these tests show, was not final salable products? we seek the answer in tests boiled away. It is probable, as pointed out above, Nos. 14 to 17. that these tests show only part of that actually reRuns 5 and 6 were made at the small mill of the tained. The small quantity left in this 96 per cent. Louisiana Sugar Experiment Station, the former sugar may justify the conclusion that in refining’ the without the addition of any formaldehyde either in sugar this excess is all eliminated. that or the immediately preceding run, and the latIf the sugar is the only product used for food, then ter with the addition of formaldehyde to the fresh there need be no hesitation, on the score of health juice in the proportion of about I part of formalin considerations, in using formaldehyde as a preserva(40 per cent. sol. of formaldehyde) to 160,000 parts tive in juice in the sugar mill. If, however, the moof juice, or 2 5 parts CH,O to a million. The clarifica- lasses is also to be used for food, or if table syrup is tion in each run was as usual with sulphurous acid, to be manufactured, then the question of the perfollowed by liming back almost to neutrality. .The missibility of its use may reasonably be raised. To concentration was accomplished as usual with re- this question we deem the data now a t hand to be duced pressure in the “effects” and vacuum pan. insufficient basis for a definite answer. We should The sugars used in these experiments (Nos. 14 and have the results from further tests of final and of IS) are the first sugars, testing about 96 per cent., intermediate products. We should develop a method and the molasses samples (Nos. 16 and 17) are of to circumvent the interference of sulphurous acid or the first molasses in each case, testing about 43 per other substanes in the distillates, in the formaldehyde cent. sucrose and 58 purity. I n the distillation with determination. We should make comparative tests superheated steam and the formaldehyde determina- also with syrup or molasses and sugar made by the tion in these samples, carried out as above described, open kettle method of boiling, with and without a peculiar condition developed which was not noted formaldehyde. We should, moreover, know in what in the previous samples with pure sugar solutions. form or combination the formaldehyde is held back The first one, or two fractions of the distillate, were in these sugar solutions, and whether it is deleterious slightly cloudy, and these same fractions, which to health in this form or combination. should have been highest in formaldehyde, gave no reaction, or only very weak ones for that substance. SOME EFFECTS OF FEEDS UPON THE PROPERThis leads us to suspect that our test in these cases TIES O F LARD. does not show the total quantities of formaldehyde B y C L HARE present. Suspecting that the interfering substance Received April 11. 1910 may be sulphurous acid, or possibly other sulphur It is well known that in the feeding of hogs differcompounds, tests were made and it was found that ent feeds produce varying effects upon the carcasses the addition of sodium sulphite strongly interfered, of the animals and upon the characteristics of the that sodium thiosulphate completely prevented the fats produced. color development, and that hydrogen sulphide but The Department of Animal Industry of the Alaslightly affected the intensity of the reaction. Not- bama Polytechnic Institute has through a series of withstanding these probable shortages in the total years made extended experiments in pork producformaldehyde recovered, we still find evidence of a n tion, particularly from an economic standpoint. excess of formaldehyde in the products from juice Working in cooperation with that department, to which it had been added, very small in the case of there have been encountered lards which show interthe 1st sugar, but more pronounced in the case of esting effects of the feeds used in their production.
I
*
I
H A R E O N EFFECTS OF FEEDS UPON T H E PROPERTIES OF LARD. The following is a preliminary account of the investigation undertaken to point out some changes in properties so brought about. Among the available pork producing feeds some of the most important economically are corn, peanuts, soy beans, cottonseed-meal and tankage, the three last being relatively rare as constituents of rations for hogs. The results here given were secured on lards resulting from three series of feeding experiments in each of which series these feeds were used. In the first series of experiments examination was made of lards from thirty-six hogs. The pigs in this series were divided into six lots and the feeding so conducted a to show the effect upon the lards of corn, cottonseed-meal, tankage, soy beans and peanuts. The lots were further subdivided so as to show the effect of corn following peanuts, of cottonseed-meal following peanuts, and of tankage following soy beans and peanuts. In the second feeding trials results were again obtained on thirty-six pigs-the feeding in this series being intended to show the effects of, 1st corn, 2nd cottonseed-meal, 3rd tankage, 4th soy beans with varying rations of corn, 5th soy beans and peanuts, 6th soy beans and peanuts with a finishing ration of corn, 7th soy beans and peanuts with a finishing ration of corn and cottonseed-meal, 8th soy beans and peanuts with a finishing ration of corn and tankage. The third experiment included twenty-three pigs and data were secured illustrating the effects of, Ist, corn; 2nd, tankage, 3rd, cottonseed-meal; qth, soy beans with varying amounts of corn; 5th, soy beans with a finishing ration of corn; 6th, soy beans with a finishing ration of corn and cottonseed-meal; and 7th, soy beans with a finishing ration of corn and tankage. Corn being the universal hog feed, one corn-fed lot was carried in each series as a basis of comparisonthe corn lards being used as standards. In every series of experiments the feeding of each lot was so conducted and the slaughtering so regulated in point of time that the full effect of the ration might be observed. Thus, in the corn-fed lot of series No. I the pigs were slaughtered a t intervalsthe first a t the end of thirty-five days and the last at the end of one hundred and forty-seven days. Slaughtering one pig from each lot a t stated intervals furnished lards on which comparable results could be secured from all pigs in the different lots as well as in the same lot. The period of feeding in days for each pig is indicated in the tables. The rations of corn, cottonseed-meal and tankage were fed soaked. The pigs were made to graze the soy beans and peanuts.
265
All pigs in each series entered the feeding trials a t a fairly uniform age and weight. All lards were rendered from the kidney fat and thus represent uniformly the firmest fat obtainable from the pig. Since the melting points and iodine values of lards are more notably affected by feeds than are the other characteristics, and a t the same time probably furnish a more accurate index to the qualities of lards than do the other constants, these values were selected as giving a sufficient showing of the changes produced. In Table I are found results on lards from hogs which had been fed on corn for periods varying from 35 to 150 days. The results include lards from three feeding trials. . The lards in series 2 , with the exception of those numbered 6 1 ~ 4 2 ~ 5 15, 3 , 23, and 34, had become slightly rancid when the iodine values were determined, but it is not thought that the oxidation had proceeded so far as to seriously affect those values. TABLEI. No. days fed. 63 91 119 140 147
pig.
Melting point.
Iodine value.
101 103 106 105 104
43.5 43.4 44.2 41.6 40.5
53.87) 58.68 57.60 } 57.37 57.10J
No. of
Series.
Ration.
I I
1
Cornmeal
2
Cornmeal
3
Cornmeal
-
Av., 5 6 . 9 2 35 70 84 112 134
72 71 73 75 74
43.1 44.8 44.0 43.0
. .. .
1
56.50 51.50 49.70 } 56.00 56.57 J
-
1
Av.. 5 4 . 0 5 75 110 125 150
69 36 42 8
:::I .... . .. .
60.041 60.07
-
Av., 6 1 . 9 8 Average, 5 7 . 3 5
Examination of these figures seems to show that corn as fed in these trials has not the effect of softening pork to the extent indicated by results secured by some investigators, notably those of Shuttl On the other hand, the iodine values are somewhat below the average value of that constant for lards-this being particularly true of the lards in series 2 , which gave an average iodine value of only 54.05. Not only do the melting points and iodine values indicate relatively firm lards throughout but the pigs which were fed corn continuously for a period of five months show no greater degree of softness than those fed for shorter periods on this ration. In fact, the pigs in series 3 show a n increasing firmness in the fat, though this may be attributed to the increasing age of the animal. Table I1 shows the effect upon the two constants under consideration of a ration containingpmuch cottonseed-meal. Canada Cent. Ex. Sta., Bull. 88.
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y .
266 No. of
TABLE 11. Iodine
No. of pig.
Melting point.
35 63 91 119 140 147
112 109 111 110 108 107
47.7 49.8 49.3 48.3 50.0 46.8
61.00) 53.57 53.57 57.58 55.16 55.95J
35 70 84 112
61 65 64 62
49.2 50.8 50.0 49.6
48.44) 51.20 51.60 50.22
days fed.
value.
Series.
i1
Average,
Cornmeal 2/3, cottonseed-meal 1/3.
'
2
53 .SO
This mixture contains a very little more oil than the corn ration but constitutes a much higher protein ration. It will be observed that the melting points are materially elevated, while the iodine values suffer a corresponding depression. This tendency has been often observed in this as well as in other rations where the protein content has been increased-e. g., corn and skim-milk, and is even more apparent in a ration containing much smaller amounts of cottonseed meal than the above. Table I11 gives results on lards produced by a ration consisting of corn 91 IO and cottonseed-meal 11 IO. TABLE111.
No. of
No. of
days fed.
pig.
70 84 112 134 99 75
45 41 44 40 42 39
Melting point.
Iodine value.
48.6
53.80)
48.4 47.6 49.0 49.7
50 6 0 1 55 0 } 52.0 41.85)
21
-
3)
Series.
I
. ...
49.23
Average,
i
Ration. Cornmeal 9/10, cotton-
seed-meal 1/10.
days fed.
pig.
Melting point.
35 63 91 119 140
131 132 136 135 133
43.1 45.6 42.5 41.6 45.1
35 70 84 112 91 91
53 54 51 52 50 55
75 110 125 150
"'
1 43 37 2
....
Iodine value.
Series.
59.08) 50.06 I 56.59 } 57.80 57.77 J
1
:!%: 1
42.2 41.4
52.40 58.801
. ...
. ... . ... .... Average,
49.85)
Ration.
1
43 .O 42.9
....
lodine value.
. . . . . . . . . . . . . .. . . . . . . Cornmeal 213 1 ' Cottonseed-meal 1/3
Cornmeal..
' '
9/10 Cornmeal Cottonseed-meal 1/10 J
Cornmeal Tankage *
9/10 l/lo
*
'
' ' ' '
' '' ' ' ' ' '''
) . . . . . ... . .
Average of
57.35
14 lards
53 '83
10 lards
52 13
5 lards
55.67
15 lards
The increasing use of soy beans as a hog feed makes necessary the determination of the effect of this ration upon the finished product and a .study of rations which may be suited to finishing hogs which have been fattened thereon. Table V I gives figures obtained on lards from pigs grazed on soy beans and a t the same time fed p a r t , rations of cornmeal: Table VI.
TABLEIV.
No. of
TABLEV. Ration.
52.13
Here we find the results on iodine Values lower than where a more concentrated cottonseed-meal ration was fed. It is, however, probable that with results on a larger number of lards from this ration more comparable figures would be secured, as there is no apparent explanation of this slight difference in properties. I n order to test further the effects of a feed containing more protein than corn, one lot of pigs in each series was fed a mixture consisting of corn 9/10 and tankage I / IO. Table IV gives analytical results on the lards produced. No. of
The figures in this table again seem to indicate that the introduction of a high protein feed into the ration produces an elevation of the melting point and depression of the iodine value of lards. While the effect upon these two constants is not so pronounced as in the case of cottonseed-meal they are nevertheless significant and would seem to point to the presumption that the changes in properties brought about in fats by cottonseed meal are, a t least in part, due to the protein content of that feed. For purposes of comparison, the averages on all lards produced from the four rations so far discussed are presented in Table V :
Ration.
1
June, 1910
Cornmeal 9/10 Tankage 1/10
I
57.02) 61.25 52.26 53.61)
1
55.67
Same
No. of
No. of
days fed.
pig.
35 75
34 76
35 35
125 23
75 35 75
87 15 68
Melting point. 33.61
? 41.8 40.0
Iodine value. 100.5 94.86
Series.
}
71.09 80.60
k}
84.32
3.
? 40.7
? Average,
-
Ration. Soy beans and 1/4 ration of corn S o y b e a n s and 1/2 ration of corn Soy beans and 3/4 ration of corn
84.72
Representing as these values do results on lards the firmest obtainable from the pigs-that rendered from the kidney fat-they illustrate the very great softening effect of this feed upon the fat. Some of the lards in this lot remain' semi-fluid throughout the winter months and are too soft for melting point determinations. Whether the softness is due solely t o the oil in the feed or in part to the fact that it is grazed as a green feed has not yet been determined. It is probable, however, that both factors influence the properties of the lard. This would be in keeping with Shutt's results on pigs grazed on clover. Further, his observations that certain other green crops, as turnips and sugar beets, have no softening effect upon lards are confirmed by data secured in this laboratory from pigs grazed on sorghum. This leads to the suggestion of the possibility that
the legumes when fed green to hogs have a softening effect upon the fat. While the soy bean feed is high in oil content, it is also relatively high in protein. The results therefore in a measure contradict the assumption that protein in the feed induces hardness in the fat. This fuither suggests the question whether the protein molecule in this feed breaks down into cleavage products differing from those formed by cleavage of the protein molecule of corn and tankage and, in turn, produces radically different effects upon the fat. In order to determine the persistence of the softness of fat produced by soy beans, three different finishing rations were employed : corn, cottonseedmeal and tankage. Table VI1 illustrates the effect of corn in hardening the fat:
was expected and the hardening effect would possibly have been greater could the time of feeding cottonseed-meal have been extended. The softness induced by the soy bean ration is somewhat in evidence after 50 days' feeding on cottonseed-meal. The effect of tankage, the third finishing ration used in these experiments, is set out in Table IX: TABLEIX.
Y o . of days fed soy beaus.
--
No. of days fed soy beans.
75 75 75 75
bean grazing and 1/4 ration of cord followed b y corn.
No. of pig. 76 98 88 7
Todine value. 94.86 83.92 69.77 70.31
__
Series.
1 3
1
Finishing ration.
Corn 35 d a i s Corn 50 days Corn 7 5 d a y s
Average of corn-fed 61 .98 lot in this series
The limited number of pigs in this lot renders the results none the less valuable, inasmuch as they are fully confirmed by the figures in Table X I , which show the effect of corn following soy beans and peanuts. Although these results show that corn produces a marked hardening effrct, yet a comparison of the iodine values on numbers 88 and 7 with the average value of the corn-fed lot of this series, indicates that the properties conferred by the soy bean ration have in some measure persisted even after a period of 75 days' feeding on corn. Some difficulty was experienced in testing the hardening effect of cottonszed-meal as in this series nearly all pigs on a ration of this feed died of cottonseedmeal poisoning. However, two cottonseed-meal pigs in this trial made good growth and Table VI11 sets forth the results secured on these : TABLEVIII. Preliminary ration soy bean grazing and 1/2 ration of corn.
No. of days
No. of
fed soy beans.
pig.
Iodine value.
Series.
75 75
87 88
82.70 73.41
3
75
56
65.46
-
Finishing ration. Cornmeal 2/3 Cottonseed-meal 1/3 for 35 days Cornmeal 2/3 Cottonseed-meal 1/3 for 50 days
Average of corn lards of this series, 61.98
The effect of the cottonseed-meal ration is what
No. of yig.
Iodine value.
6s
84 3 2
5;
68.83
75
66
72.67
75
77
74.0
13
_. I5
TABLETI1 Ration-Soy
bean grazing and a 3/4 ration of corn
Preliminary ration-Soy
Average values corn lards,
Series. 3
Finishing ration. Cornmeal 2/3 Tankage 1/3 for 35 days Cornmeal 2/3 Tankage 113 for 50 days Same 75 days
61.98
Here again there is illustrated the pronounced hardening effect of a high protein ration. For some unexplained cause the hardening effect in this trial with tankage seems to be greatest a t the end of 35 days' feeding period and the lards to increase in softness for longer periods of feeding. This is probably due to individual peculiarities of the pigs. There still appears a persistence of the. properties originally contributed by soy beans. A study was also made of the persistence of the properties conferred upon lards by a ration of soy beans followed by peanut grazing. There is first given in Table X values obtained from pigs slaughtered off peanuts following soy beans : TABLEX. Ration-Soy
bean grazing followed b y peanut grazing.
KO.of
days fed soy beans.
35
No. of pig. 127
Melting point.
Iodine value.
Series.
Finishing ration.
40.6
79.61
1
Peanuts28 days
25
31.2?
89.84
2
Peanutsand 1/2rntion corn 35 days
35
14
37.0
76.91
2
Same
35
33
?
89.70
2
Same
35
Average,
84.02
A comparison of these lards with those from the soy bean fed pigs, as shown in Table VI, demonstrates that while the average results of the two sets are practically the same, yet some of the individual soy bean lards are very materially softer than the softest of this lot, and that as between soy beans and peanuts the former has probably the greater softening effect. In efforts to harden the fat of the pigs fed upon both soy beans and peanuts, use was again made of corn, cottonseed-meal and tankage. The effect of corn is shown in Table XI:
268
T H E JOURNAL OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y .
Ration-Soy No. days * fed soy beans and No. of peanuts. pig.
beans followed b y peanuts followed by corn.
TABLEX I .
70 70 70 70
14
10 12 11
Melting point.
Iodine value.
37.0 36.9 36.8 39.0
76.94 82.80 73.93 70.1
Series. 2
Finishing ration. '
Corn 14days Corn42 days Corn 64 days
2 2 2
There is found here a confirmation of the results recorded in Table VII, which illustrate the hardening effect of corn as well as the difficulty of hardening the fats to a normal firmness-using the corn lards as standard. The effect of the corn is progressive and is greatest as in Table VI1 during the first forty to fifty days. Still further confirmatory results on the effect of corn were secured from a third trial with four pigs. The results are recorded in Table XI1 : TABLEXII. Ration-Sorghum and 1/2 corn 35 days. (preliminary). No. days fed Iodine preliminary No. of Melting ration. pig. point, value. 63 63 63 63
40.8 41.8 42.3 42.7
113 116
117 118
Peanuts and 1/2 corn 28 days
70.08 62.15 60.67 60.60
Series. 1 1 1
1
I n Table XI11 are given results secured on the corn soy bean peanut pigs, finished on a ration of and cottonseed-meal. TABLEXIII. Days fed soy beans and pea- No. of pig. nuts.
'
Melting point.
Iodine value.
70
25
31.2
89.84
70
22
42.4
76.1
70
21
41.5
79.22
Series. 1
Finishing ration
2
2/3 corn and 1/3
2
Cottonseed-meal for 4 2 days Same for 64 days
These results are not so striking as those found in Table VIII, but the one set is confirmatory of the other. Cottonseed-meal in the ration exercises its usual function of firming the fat. The lot in this series finished on tankage puts this feed in a most favorable light as a finishing ration, as will be observed by inspection of the figures in the following table : TABLEXIV. beans 35 days followed by peanuts 35 days.
.
Preliminary ration-Soy No. days fed soy beans and No. of Meiting peanuts. pig. point. 70 70 70 63 63 63,
....
33 35 31 127 126 130
,
39 4 1 .O
40.6 41.9 42.9
Iodine value.
Series.
89.70
2
74.35 70.65 79.61 66.40 64.69
) 2
;) 1
Finishing ration. Corn and 1/3 tankage 42 days Same49 days Corn and 1/6 tankage28 days Same 77 days
While, as before stated, these results point to tankage as a desirable ration for hardening the fat of soft hogs yet, on considering them in connection
1910
with the results in Table IX, it is seen that confirmatory feeding trials are necessary before the results in Table XIV can be accepted. Tankage, however, is free from the. toxic effect which renders cottonseed-meal a dangerous hog feed and the data secured are sufficient to emphasize its value as a finishing ration. The results too show that it compares favorably with cottonseed-meal in its hardening effect on the fats. This, however, is not verified by examination of the carcasses nor by the results on pigs fed these rations without a prior softening ration-see Table V-and further comparative feeding trials are in progress. Below is given a further comparison of lards from the two feeds: TABLEXV. Preliminary ration-Soy beans 35 days. Days fed soy beans, peanuts and No. of Melting Iodine corn. pig. point. value.
Peanuts 35 days. Corn 14 days
Series.
84 84
32 10
36.9 39.2
82.80 71.60
2 2
84
20
44.4
71.80
2
Finishing ration. Corn 2/3, tankage 1/3 for 14 days. Corn 2/3, Cottonseed - m e a 1 1/3 for 14 days.
Finishing ration. Corn28 days Corn 56 days Corn 77 days
June,
The results on this lot of three pigs but serve to further confirm results recorded in the foregoing tables illustrative of favorable changes produced in the properties of lards by these two rations. Summarizing the results, it is found that corn may be fed without regard to time limit in so far as concerns any unfavorable effect on the properties of the lard. Soy beans and peanuts both produce oiliness in the hogs and the lards resulting are extremely soft. The fat of oily hogs can be hardened by rations of corn, cottonseed-meal or tankage-the two latter producing unusually firm lards. The possibility is suggested that legumes when fed green may produce soft fats. In connection with the general subject here discussed, there are presented below results which accentuate the extreme differences in properties which may be conferred upon lards by different feeds:
34
Melting point. ?
100
48.2
No. of
pig.
Iodine value. 100.5 40.35
Iodine value of liquid Saponification value. fatty acids. 119 78
197.5 203.2
Pig No. 34 is selected from the soy bean fed lot. No. IOO was fed throughout its life on slaughter pen refuse and weighed about 2 0 0 pounds when slaughtered. This lard presents much the appearance of beef tallow and its constants possess much the same value as those of beef fat. It further responds readily to the Belfield test for beef fat. ALABAMA POLYTECHNIC INSTITUTE.
.
